Pathology Visual Question Answering on PathMMU tiny (test)
79.6Overall ScorePathoSage
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| PathoSageHost Model=GPT-5.42026.05 | 79.6 | 79.4 | 78.9 | 82.8 | 83.7 | 71.2 | |
| Gemini-3-Pro-Preview2026.05 | 78.6 | 79.4 | 76.6 | 80.8 | 86.1 | 67.8 | |
| Gemini-3-Flash-Preview2026.05 | 78.1 | 79.1 | 74.3 | 81.2 | 85.6 | 65.5 | |
| Patho-AgenticRAG2026.05 | 73.2 | 72.2 | 74.7 | 76.5 | 79.3 | 57.2 | |
| GPT-5.42026.05 | 72.9 | 76.5 | 73.4 | 79.6 | 75.5 | 53.7 | |
| CPath-Omni2026.05 | 72.4 | 74 | 76.6 | 69.8 | 65.9 | 75.7 | |
| GPT-5.4-mini2026.05 | 70.1 | 73 | 72.9 | 79.6 | 69.2 | 53.7 | |
| Patho-R12026.05 | 69.5 | 72.2 | 67.9 | 75.3 | 81.7 | 44.6 | |
| Qwen3-VL-32B-Instruct2026.05 | 65.4 | 71.5 | 67.9 | 69 | 63 | 50.3 | |
| PathGen-LLaVA2026.05 | 60.1 | 60.1 | 60.9 | 60.8 | 63.5 | 54.2 | |
| HuatuoGPT-7B2026.05 | 58.2 | 61.9 | 58.7 | 60 | 65.9 | 40.1 | |
| Qwen3-VL-8B-Instruct2026.05 | 55.9 | 61.9 | 60.6 | 55.3 | 54.8 | 42.9 | |
| InternVL3-8B2026.05 | 55.4 | 57.7 | 60.6 | 54.9 | 58.2 | 42.9 | |
| InternVL2.5-8B2026.05 | 50.1 | 55.9 | 57.8 | 50.6 | 51.4 | 29.4 | |
| Quilt-LLaVA2026.05 | 45.6 | 47.3 | 46.4 | 51.8 | 46.2 | 32.2 | |
| LLaVA-1.5-13B2026.05 | 38.8 | 44.5 | 40.4 | 34.1 | 47.1 | 24.9 | |
| LLaVA-Onevision-7B2026.05 | 36.9 | 37.7 | 35.7 | 47.1 | 38.9 | 20.3 | |
| InstructBLIP-FLAN-T5-XXL2026.05 | 34.3 | 39.1 | 33.6 | 34.5 | 38.5 | 22.6 | |
| LLaVA-Med2026.05 | 25.3 | 28.5 | 28.9 | 22.7 | 22.6 | 22.6 |